November 2023 Summaries

Hasura has announced the alpha version of its Data Delivery Network (DDN), a platform designed to address challenges faced by developers in building APIs and microservices. The key features of this new platform include language agnosticism, enabling developers to create complex API features quickly, and offering a free and intuitive developer environment with robust tooling for efficient API design, testing, iteration, and maintenance. The Hasura DDN alpha release includes features such as a new console geared for rapid onboarding and troubleshooting, code-first declarative API authoring and iteration, feature-rich data domain modeling, embedding business logic using TypeScript, and native extensibility with data connectors.

This article debunks common myths and misconceptions about GraphQL, a query language used in web development. The first myth is that GraphQL allows clients to perform arbitrary queries, which is true during the development phase but restricted in production with allow lists or persisted queries. The second myth is that GraphQL is similar to SQL for graph databases, but it's actually designed to model and execute semantics as a tree structure. The third myth is that GraphQL servers are hard to build and maintain, which can be resolved by using engines like Hasura to automatically build the API based on data modeling. The fourth myth is that GraphQL has the N+1 query problem, but this issue can be mitigated with dataloader batching or efficient tools like Hasura. The fifth myth is that GraphQL is unnecessarily complex for small teams, but it becomes more valuable as a project grows and requires richly connected data across different sources. Lastly, the article addresses the concern about caching in GraphQL, stating that while it can be tricky, proper client and server-side caching strategies can be implemented with existing tooling.

Remote Table Authorization is a new feature that allows users to define write permission rules across all their data. This innovative project was born out of the need for complex access control in data management. With this feature, users can now use data from different sources to perform authorization checks on any table. The permission system has been enhanced to filter based on fields from remote relationships, making it easier to manage and grant access to distributed data units across various systems. Despite some limitations, Remote Table Authorization offers a robust solution for complex data access scenarios and enhances the overall functionality of GraphQL APIs.

The text discusses the adoption of GraphQL and its comparison to REST. It highlights that while GraphQL done right is better than REST in terms of efficiency and speed, it's hard to implement correctly. The author suggests that companies with great communication structures are more likely to succeed in adopting GraphQL due to Conway's law. They also mention the possibility of using both GraphQL and REST together. The text then delves into personal experience with building GraphQL APIs, stating that it is difficult and requires a different mindset from REST API design. It highlights common mistakes made by developers when adopting GraphQL and emphasizes the importance of training teams on how to use GraphQL correctly. The author also talks about the benefits of consuming well-thought-out GraphQL APIs, which can be great for users once they are built. The text concludes with a discussion on supergraphs, an architecture and operating model that combines the best aspects of monolithic and microservices approaches. It suggests thinking like a node when building a supergraph and emphasizes the importance of having a well-defined ubiquitous language within an organization. The author invites readers to share their experiences and thoughts on GraphQL and supergraphs, fostering a community of learning around these technologies.

The blog discusses how GraphQL can be used as an interface for data products. It explains that REST APIs are a well-established mature ecosystem for security and monitoring but have lagged behind in adoption due to their complexity. In contrast, GraphQL has demonstrated its efficiencies and provides benefits over traditional RESTful approaches such as reducing over fetching and under fetching of data, providing a standard for data access, and offering flexibility and higher control over the data received from API calls. The blog also explores challenges and opportunities with using GraphQL, including its adoption lagging behind REST APIs, but easing infrastructure framework creation through companies like Hasura. It delves into how GraphQL can be used to improve user experience, architecture, performance, security, cost, and use cases for data products. The blog concludes by stating that it is time to explore the next generation of API access via GraphQL, which is better suited for AI applications.

The text discusses a method to convert TypeScript types into GraphQL types for better serverless functions. By using the Hasura v3 connector architecture and parsing TypeScript code, developers can create a fully typed API in GraphQL. This approach allows for more efficient function composition and building semantic graphs compared to traditional HTTP endpoints. The goal is to empower TypeScript developers to deploy their work and make it usable for a wide audience while incorporating functions into larger organizations' supergraph implementations.